Microsoft's Microsoft-Decision-1 is a specialized decision-scoring model for engineering teams building routing, classification, and verification logic into their AI systems. Announced on October 9, 2026, it returns calibrated scores for a fixed set of choices rather than generating free-form text, which makes it predictable to drop into production pipelines. It targets teams that already run model routers and want a cheaper, faster scoring step than a full LLM call.
Core Features
- Produces calibrated scores across a fixed list of options for routing and classification tasks.
- Supports prioritization, where it ranks candidates by a learned preference signal.
- Handles verification-style checks that confirm whether an output meets a rule.
- Available through Microsoft Foundry and OpenRouter, so it slots into existing model routers.
- Designed for low-latency scoring rather than long-form generation.
Use Cases
- Agent platforms that need to route a request to the right sub-model or tool.
- Triage systems that prioritize support tickets or leads by urgency.
- Quality gates that verify whether a generated answer should be released.
Pricing
Priced as a hosted inference model on Microsoft Foundry and OpenRouter; exact per-token rates follow each platform's model pricing. There is no standalone free tier beyond platform trial credits.
Our Take
Best for builders who want a dependable scoring layer instead of prompting a general LLM to decide; the trade-off is that it is narrow by design and not a substitute for a reasoning model. See related models in our AI Engine/Model category.
